The use of artificial intelligence in image generation is huge. Stable Diffusion is an open source image synthesis model that can be used by anyone with a PC and a good graphics card. If you give it a descriptive phrase, the results appear on your screen like magic.
Some artists are delighted by the prospect, others aren't happy about it, and society at large still seems largely unaware of the rapidly evolving tech revolution taking place through communities on social media The creation of visual art itself, as well as the invention of the camera, bring implications similar to image synthesis. Our sense of history may be at stake. Stable Diffusion leads a new wave of deep learning creative tools that are poised to change the way visual media is created.
Stable Diffusion is the brainchild of Emad Mostaque, a London-based former hedge fund manager who wants to bring novel applications of deep learning to the mass. Stable Diffusion was not the first image synthesis model to make waves this year.
DALL-E 2 shocked social media with its ability to transform a scene written in words into a variety of visual styles. astronauts on horseback, teddy bears buying bread in ancient Egypt, and novel sculptures in the style of famous artists were created by people with privileged access to the closed off tool.
Advertisement
After DALL-E 2, the two companies announced their own text-to- image models. MidJourney is open to the public a few months later and charges for access but has a more illustrative and artistic quality.
Stable Diffusion is next. The open source image generation model released by Stability is similar to DALL-E 2. DreamStudio is a website that sells access to compute time for generating images. Stable Diffusion is open source and anyone can use it.
Dozens of projects that take Stable Diffusion in new directions have sprung up in the last week. People have achieved unexpected results using a technique called "img2img" that has "upgraded" MS-DOS game art, converted a scene from Aladdin into 3D, and more. The ability to richly visualize ideas to a mass audience may be brought about by image synthesis, similar to what Adobe did in the 1990s.
If you follow a few steps, you can run Stable Diffusion yourself. For the past two weeks, we've been running it on a Windows PC with an Intel Core i9-9900K processor. In about 10 seconds, it can create 512512 images. The time on a 3090 Ti goes down to four seconds. The interface is getting more and more complex, going from crude command-line interface to more polished front-end GUIs. If you aren't technically inclined, hold onto your hats, easier solutions are on the way. You can try a demo online if you don't succeed.
All Rights Reserved © 2024
